Skip to main content

Table 4 Model calibration of breast cancer-specific mortality regressed on prognostic variables in SEER data

From: Development of a prediction model for breast cancer based on the national cancer registry in Taiwan

 

Breast cancer-specific survival

Overall survival

Calibration year

Observed

Predicted

P value

Observed

Predicted

P value

Training data

 1

51

52.64

0.821

83

83.87

0.924

 2

229

236.75

0.615

316

317.83

0.918

 3

363

373.45

0.589

481

475.71

0.808

 4

425

422.94

0.920

569

544.60

0.296

 5

348

350.04

0.913

466

456.30

0.650

 6

220

215.38

0.753

290

275.41

0.379

Testing data

 1

33

25.28

0.125

47

41.21

0.367

 2

109

114.89

0.583

142

157.12

0.228

 3

177

182.81

0.668

237

237.48

0.975

 4

202

206.35

0.762

266

270.31

0.793

 5

176

172.28

0.777

227

226.75

0.987

 6

96

104.32

0.416

129

133.50

0.697

White

 1

236

132.35

< 0.001

416

245.11

< 0.001

 2

472

432.07

0.055

805

666.75

< 0.001

 3

554

570.83

0.481

905

834.99

0.015

 4

378

426.84

0.018

617

628.31

0.652

 5

 6

Black

 1

47

29.32

0.001

87

43.98

< 0.001

 2

132

90.05

< 0.001

194

115.66

< 0.001

 3

138

115.04

0.032

195

141.00

< 0.001

 4

99

86.07

0.164

146

106.57

< 0.001

 5

 6

Asian

 1

10

7.57

0.377

23

13.93

0.015

 2

22

24.51

0.612

43

38.03

0.421

 3

21

29.84

0.106

43

44.16

0.862

 4

17

23.31

0.191

29

33.42

0.445

 5

 6